6-lesson course · about 2 hours
Checking That Your Citations Actually Support Your Claims
For Creators, educators, influencers and personal-brand teams responsible for useful content, audience trust and a consistent production practice
Everything for CreatorsFor creators and educators who use AI research: implement a repeatable routine to extract supporting passages, flag citation mismatches, and verify every claim across all your platforms before publishing.
What you’ll be able to do
By the end, you can…
How it works
A tutor that meets you where you are
- Adaptive lessonsThe course material is not static. It monitors your progress and adjusts the complexity of the lessons to match your current skill level as you move forward.
- One-on-one AI tutorYou have a dedicated AI tutor available at all times. You can ask it questions, request examples, or clarify verification routines using either typed text or your voice.
- Active recall toolsEach lesson includes targeted quizzes and digital flashcards. These tools are tied directly to the material to ensure the verification routine becomes a permanent habit.
- Tracked progressThe platform logs your completed modules and mastery of specific outcomes. Once you finish the course, you receive a certificate of completion to mark your achievement.
Curriculum
The curriculum
Six lessons designed to move you from manual checking to an AI-assisted verification routine.
Module 1The Verification Standard2 lessons
1The Research-Assistant Standard
- Distinguish between verified factual information and personal lived experience in a script
- Apply Cleo Abram's research-assistant standard to a set of claims
2Building Claim-to-Source Packets
- Create a claim-to-source packet containing links, dates, and exact supported statements
Module 2AI-Powered Verification2 lessons
3AI Passage Extraction and Mismatch Flagging
- Use AI to extract the specific passage from a source that supports a claim
- Identify and flag mismatches where a citation is present but the source text does not support the claim
4Handling Contradictory Sources
- Set up a stop-rule to pause production when a new source contradicts an existing claim
Module 3Cross-Platform Maintenance2 lessons
5The Correction Audit
- Verify a corrected fact across all content versions, including captions, pinned comments, and spoken lines
6Verifying the Final Release
- Execute a final release check to ensure no citation appears next to a claim it doesn't support
Who it’s for
Made for you if…
Made for: Creators, educators, influencers and personal-brand teams responsible for useful content, audience trust and a consistent production practice
Creators using AI for research
You use AI to speed up your scripting and research, but you've experienced the 'citation gap' where a link doesn't actually support the claim. You need a way to verify AI-generated citations before they reach your audience and erode your trust.
Educators publishing across platforms
You distribute the same factual information across TikTok, Instagram, and YouTube. You need a repeatable routine to ensure a corrected fact is updated in every location, including pinned comments and spoken lines, so you don't provide conflicting data.
Research-driven content leads
You manage a production flow where multiple people or tools contribute to a script. You need a standardized 'claim-to-source' packet that separates verified factual data from personal observations to keep your final release accurate.
Knowledge creators scaling their output
As you produce more content, manual checking becomes a bottleneck. You want to implement AI-powered passage extraction and mismatch flagging to maintain high accuracy without spending hours hunting through source documents.
Questions
Good to know
What is the 'citation gap' and why does it happen?
The citation gap occurs when you place a source link under a claim, but the linked page does not actually support that specific statement. This often happens when creators rely on AI research without a verification routine, leading to audience members calling out the mismatch in the comments.
How does AI help with verification if it is the cause of the problem?
Instead of using AI to find sources, you use it for passage extraction and mismatch flagging. The AI pulls the exact quote from your source that supports your claim, allowing you to quickly confirm a match or identify where the source fails to support the statement.
What is a claim-to-source packet?
It is a structured record that contains your specific claims alongside their corresponding links, the dates the information was accessed, and the exact supported statements from the source. This creates a transparent evidence trail for every fact in your script.
What happens if I find a new source that contradicts my existing content?
The course teaches you to implement a 'stop-rule.' This is a production boundary that requires you to pause all related content creation immediately when a contradiction is found, ensuring you resolve the truth before publishing further.
Does this course cover how to handle corrections after a video is already live?
Yes. You will learn a correction audit routine to verify a changed fact across every version of your content, including spoken lines, pinned comments, and captions, so the correction is consistent across all platforms.
How do I know when to use a source versus when to rely on my own experience?
You will learn to apply the 'research-assistant standard' to distinguish between verified factual information and personal lived experience. This ensures you know exactly which lines in your script require a citation and which are your own unique observations.
Will I learn how to use specific AI tools for this process?
The course focuses on the repeatable routine of extraction and flagging. You will learn how to direct AI to extract specific passages and flag mismatches so that no citation appears next to a claim it does not support.